Solution-processing offers a pathway to fabricate complex chalcogenide structures for electronic, imaging and photonic applications. More importantly, it has benefits in flexibility and integration as compared to conventional methods. This paper reviews the fundamental physics and chemistry in the solution process, along with discussing recent examples of applying the solution method to fabricate high-quality optical and photonic components.
©2013 Optical Society of America
Chalcogenide glasses are well-known for their desirable optical properties such as transmission in the infrared, high refractive index and high nonlinearity. Such properties have enabled many chalcogenide-based applications to be developed in diverse fields including photonics, medicine, environmental sensing and security [1–3]. Conventional methods for depositing these materials such as vacuum coating (thermal evaporation, chemical vapor deposition or sputtering) or pulse laser deposition [3–12] have been used extensively, and when combined with etching or other photolithography steps, these approaches are efficient for fabricating 2D structures on flat surfaces. However, they can become unpractical for producing structures with complicated shapes or special dimensions. In such situations, solution-based methods, which involve the dissolution of chalcogenide glasses in solvents and processing as a liquid, can produce materials of similar optical, chemical and physical properties, while having advantages in process flexibility and device integration .
Solution-based processing of chalcogenide glasses for optical applications was first explored by Chern et al. in the early 1980s. Their goal was to fabricate thin photosensitive chalcogenide glass films to be used as photoresists in lithography in a similar fashion as standard polymeric materials [14–22]. Such an approach was met with moderate success however it did not find widespread adoption due to the difficulties in material handling. Moving beyond photoresists, other researchers were motivated by the simplicity and low cost of solution-based methods  and the resulting chemically homogeneous films as compared to traditional approaches [24, 25].
More recently, the solution-process has attracted increasing interests in the optics and photonics community, primarily for applications in the IR, which form the key part of chemical sensors, data storage devices and IR detectors. This approach stands out due to its abilities to create thick films of material, to directly integrate different components and to deposit materials on non-planar surfaces. In this article, we review the work done in solution-processing chalcogenides, starting with analyzing the physics and chemistry of chalcogenide dissolution and solution-processed films, followed by discussing individual applications employing the solution method for fabrication.
2. Chemistry of solution processing chalcogenides
2.1 Dissolution mechanisms
Understanding the mechanism of chalcogenide dissolution is key to revealing the underlying science behind the solution process and has multiple important implications. First, the study of glass dissolution is essential for explaining properties of the solution and the corresponding solution-processed films. Second, such understanding in fundamental science sheds light on improving existing fabrication techniques and expanding its range of materials. As such, this section explains the fundamental dissolution chemistry of arsenic sulfide in different solvents, as well as other chalcogenides in comparison.
Chern et al. investigated the dissolution of arsenic sulfide in n-propylamine and n-butylamine [14–19]. They treated the starting material, amorphous arsenic sulfide, as having a layer-like structure and suggested that the solvent breaks up the bulk into small flat clusters starting at defect sites between the layers. This occurs by nucleophilic substitution of a sulfide atom by the alkyl amine group of the solvent. The resulting arsenic alkyl ammonium group splits off its excess hydrogen to form RNH3+ group, which then bonds to the negatively charged sulfur dangling bond. The arsenic atom in the alkyl ammonium group now exhibits a lower electron density than prior to the substitution, with the highly electronegative nitrogen pulling away electron density from the arsenic. This makes the arsenic vulnerable to a second and third nucleophilic attack, resulting in an alkyl amino arsenic compound and three hydrogen terminated sulfide groups. The hydrogen sulfide groups in combination with additional solvent molecules are in chemical equilibrium with the alkyl ammonium sulfide salt. Breaking up the arsenic sulfide layers leads to the formation of amorphous, arsenic-deficient arsenic sulfide fragments that are terminated by excess sulfide dangling bonds and charge compensating alkyl ammonium ions (Fig. 1), as well as an insoluble alkyl amino arsenic compound which precipitates. The size (as well as the shape) of the arsenic sulfide fragments determines the extent of the sulfur excess, and it is in the order of 2-10 nm. Indeed, 25 years later, Kohoutek et al. found a correlation between solution concentration and cluster size of arsenic sulfide dissolved in n-butylamine . The polar alkyl ammonium sulfide shell around the arsenic sulfide fragments makes the material more soluble in the chosen solvent as well as in a wide range of other polar solvents. As for the precipitating alkyl amino arsenic compound the exact composition or even the chemical formula could not be determined, since it is likely to be a mix of different reaction products, including oxide-species from a competing nucleophilic attack by trace amounts of water in the solvent. However, the thermodynamic (ideally, in chemical equilibrium) dissolution products of arsenic sulfide in n-propylamine would be as follows [16, 21]:
In the late 1980s, Guiton and Pantano investigated the dissolution of arsenic sulfide in a different kind of amine solvent, ethylenediamine (EDA) [24, 25], for a low-temperature route to optical chalcogenide glass films. They ruled out the formation of alkyl ammonium salts or hydrogen sulfide in this system, but claimed the existence of polymer-like chains of As4S4 rings interlinked by bridging sulfur atoms, each ring chelated by two solvent molecules. This completely different mechanism is only possible due to the chelating nature of a diamine solvent. During drying of the solution the material loses solvent and condenses in analogy to the sol-gel process of silicate gels (Fig. 2). Guiton et al. argued that a proton source (other than the amino group) was needed in order to create alkyl ammonium species which was the reason that they were unable to detect any in their water free solutions. However, it should be noted that Chern et al. detected alkyl ammonium salts even though their suggested dissolution mechanism involves no water molecules .
A third group of researchers including Mamedov and Michailov investigated the dissolution mechanism and kinetics of arsenic sulfide glass in alkaline water solutions and amine based solvents starting in the early 1990s [20, 21, 27, 28]. They found that the dissolution mechanism in alkaline water solutions as well as some alkyl amine solvents including n-butylamine consists of two steps, first the adsorption of nucleophiles (hydroxyl anions or amine groups) at the glass surface and then the nucleophilic substitution of a sulfur atom by the nucleophile :
Thus breaking up the glass network into fragments and creating a sulfide dangling bond whose charge is compensated by either the metal cation from the inorganic base, or alkyl ammonium, as well as an arsenic hydroxide or alkyl amino arsenic species [21, 27]. This second step, the nucleophilic attack of the arsenic in the covalent glass network is consistent with the dissolution mechanism introduced by Chern et al. .
While the dissolution mechanism described here is valid for stoichiometric and chemically ordered arsenic sulfide such as bulk arsenic sulfide glass or annealed evaporated films, the dissolution rate decreases when homopolar bonds as in as-evaporated films are present, because these defects decrease the electrophilicity of the glass network . If many homopolar bonds are present, the use of amine reactants in an aprotic solvent, which allows sulfur to dissolve, offers a different dissolution mechanism that enables higher dissolution rates allowing amines to act as catalysts in forming free sulfur radicals as suggested in  with diethylamine as the solvent:
In addition, Mikhailov et al. suggested the effect of material stoichiometry on dissolution rate. In particular, the dissolution rate of the film in the dipentylamine solution is found to depend on the film composition, excess sulfur concentration in the solution and the spatial orientation of the sample. This could be due to the different reactivities of sulfur, as the intermediate product of the dissolution process .
The detailed study of As2S3 has inspired research on other chalcogenides, and one of the them is As2Se3, which shares a similar As-chalcogen trigonal pyramidal structure. Although arsenic selenide dissolved poorly in propylamine and butylamine, Zou et al. found that up to 0.6g/mL As2Se3 could be dissolved in EDA with no clear phase separation. Raman study shows peaks corresponding to As4Se4 unit, Se-Se chain, Se-Se ring and AsSe3 pyramidal unit. Similar to As2S3 dissolution in EDA, the first three units become present in the dissolution process as As-Se heteropolar bonds breaks to form Se-Se homopolar bonds .
More complicated ternary chalcogenides have also been explored with solution process. In a study of As33S67-xSex (x = 0, 17, 33.5, 50 and 67 at.%) processed with butylamine, similarities were found with respect to previously-studied chalcogenides. Raman spectrum showed peaks of AsSe3 pyramids, Se-Se chains, AsS3 pyramids, As4S4 units, weak S8-rings and S-Se bonds in spin-coated and pre-baked films. It is observed that these bonds approach bulk values along with solvent removal, except for the fragmentation and disappearance of S and Se ring structures .
Addditionally, Waldmann et al. studied the dissolution chemistry of a Ge-based chalcogenide glass with the composition Ge23Sb7S70. The Ge building blocks of this glass, GeS4-tetrahedra, rearrange during dissolution in n-propylamine to form adamantane-like Ge4S104- clusters . This molecular structure is retained during spin-coating, leading to thin chalcogenide films which are more similar to the solution than the bulk glass. Raman spectrum measurement shows the disappearance of solvent peaks and band shifts indicating the formation of a tighter structure. In another comparative, it was found that Ge23Sb7S70 glass dissolves with a similar rate as As2S3 when immersed in KOH, but had a higher rate when immersed in NH4OH and a lower rate in propylamine or butylamine. One of the causes for the higher rate could be the difference in material structures and dissolution routes. Another factor could be the less covalent nature of Ge-S bonds than As-S bonds, and therefore Ge23Sb7S70 should dissolve faster in more polar solvents such as OH-based solutions and slower in organic solvents such as amines .
Furthermore, Orava and Kohoutek investigated the dissolution of As-S, As-Se-S and As-Se thin films in inorganic aqueous bases as well as amines. They confirmed the existence of different dissolution mechanisms, resulting in varying etching rates and chemical selectivity [22, 26, 33, 34]. The same group also demonstrated solution-processing of ternary and quaternary chalcogenides by first introducing Ag to the As-S, As-Se-S and As-Sb-S system through melting [34–37]. The rate of dissolution was found to depend on film composition film, choice of etchant and its concentration, with the highest resolution on the scale of nanometers.
Although each solvent system behaves differently with a chalcogenide solute, there exists a common pattern in the basic dissolution mechanism. The solvent molecules tend to be electron-rich and can initiate dissolution with nucleophilic attacks. A similar dissolution process is expected for other chalcogenide materials with appropriate solvents.
2.2 Solution properties
Dissolution is usually the very first step of solution processes. Finely grounded amorphous or crystalline  arsenic sulfide readily dissolves in different amine solvents, such as n-propylamine, n-butylamine and EDA. Typically they are mixed together and stirred for a few days, preferably in absence of moisture and air. Solutions can be very highly concentrated: solutions up to 0.4 g/ml in n-propylamine  and 1.5 g/ml in EDA  have been fabricated. A different route was taken by Santiago et al. who also dissolved chalcogenide glasses in n-butylamine, but then dried the resultant solutions to obtain the butyl ammonium salt as a powder which they subsequently dissolved in N,N-dimethylacetamide. The amide solution is less hygroscopic and can be used to spin-coat multiple layers .
The concentration and viscosity affect the flow properties of the solutions during processing. Spin-coating of denser solutions leads to thicker films, and homogenous films in a wide thickness ranging between tens of nanometers and tens of microns can be fabricated . Taking it further, thicker films over tens of microns was recently demonstrated and detailed in section 4.2.
The color of chalcogenide solutions ranges between pale yellow and dark amber in arsenic sulfide and selenide solutions, depending on concentration and type of solute . The solution viscosity of arsenic sulfide in EDA as well as n-propylamine shows Newtonian behavior and increases with concentration [25, 32]. The solution properties of arsenic sulfide do not depend on whether crystalline or amorphous material is used , and the solutions can be stored for months under inert gas atmosphere. Occasionally, slowly-formed precipitation may occur due to oxidation and can be filtered out prior to processing.
3. Properties of solution-processed films
Spin-coating dissolved chalcogenide glasses is one of the most common techniques used in solution processes. It is usually done in the dry, oxygen-free atmosphere of a glovebox to prevent reaction with moisture or oxygen from the air. A few drops of the solution are deposited onto a substrate which is then spun at different speeds to produce films of different thicknesses. Since there are no sudden temperature changes involved in solution-based spin-coating, different thermal expansion properties of substrate and film material, which have to be considered for conventional deposition techniques, are usually not an issue.
Spin-coated films are usually treated thermally in two steps, a soft-bake at lower temperatures that removes most of the solvent, and a hard-bake closer to the glass transition temperature of the bulk glass to remove residual solvent and reaction products, and to densify the glass. For instance As2S3 films made from amine solutions start to lose molecular amine at 70 - 80°C when the alkyl ammonium salts decompose, leaving behind HS-terminated clusters. At 130-150°C, these clusters start to interlink, releasing H2S and leaving behind a material similar to the bulk according to the reaction,13, 15]. The stoichiometry of the material is maintained because the S evolved with H2S is compensated by the As precipitation mentioned in section 2.1. Measurements also show that the resulting film stoichiometry approaches that of bulk with an increasing annealing temperature .
3.1 Chemical structures and light exposure
Due to the different nature of the deposition techniques, vacuum deposition vs. solution methods, the deposited materials show differences in structure, and thus they respond differently to thermal treatment and exposure to light. Arsenic sulfide which has been evaporated at high temperatures and under vacuum is known to form As4S4 clusters and sulfur chains which can be found in the deposited films. The homopolar bonds (As-As in As4S4 and S-S in Sn) are susceptible to light and easily break under light exposure as well as thermal treatment, leading to a material closer to arsenic sulfide bulk glass. On the other hand, arsenic sulfide dissolved in alkyl amines, as explained in section 2.1, can be described as a nanocolloidal solution consisting of flat clusters, that internally retain the structure of the layer-like starting material, but capped by ionic pairs of sulfide dangling bonds and alkyl ammonium molecules. These nanoclusters agglomerate in spin-coating forming larger clusters . During baking the material first loses non-bonded solvent molecules before the alkyl ammonium sulfide groups split off alkyl amine molecules under the formation of hydrogenated arsenic sulfide at around 80°C. Above 130°C, the hydrogenated sulfide groups evaporate as H2S , leaving behind the stoichiometric arsenic sulfide As2S3 with a structure close to the bulk material as described above. This can be confirmed by the converging refractive index and infrared absorption spectrum with the annealing duration. Comparing 1 hr baking at 60°C with higher temperatures up to 180°C, the optical bandgap of resulting thin films decreases to 2.23, refractive index increases to 2.4 at 900nm, density increases to 3.8g/cm3 and the solvent absorption peak decreases significantly giving transmission over 80% .
Spin-coated films show qualitative agreement with evaporated films with respect to their photo-induced phenomena. For instance, the response of solution-deposited arsenic sulfide to light shows signs of approaching the bulk material, in terms of an increase of the refractive index as well as densification . In other studies, researchers found photodarkening and photo-induced Ag diffusion in spin-coated As-S, As-Se and Sb-S films which were consistent with evaporated materials [32, 39–42]. Furthermore, photo-induced linear dichroism was shown in spin-coated arsenic selenide films, with a value even larger than evaporated films , which could be associated with the larger “free volume” in spin-coated films.
3.2 Physical structures
The surface morphology of spin-coated films is influenced by solvent evaporation during fabrication, therefore is affected by solvent properties and preparation condition. In a study of spin-coating As33S67 with butylamine solutions, grain sizes of about 25 and 50nm were found on films produced by a dilute solution and such grainy pattern flattened out after thermal treatment. The author concluded that this grain structure was from the agglomeration of 1 to 4.2 nm radii clusters during solvent evaporation . In another example, the film properties of spin-coated Ge23Sb7S70 films are investigated against spinning parameters such as spin-speed and wait time between applying the solution to the substrate and spin start. Increased spin speed leads to thinner films with lower surface roughness. Another important result from this study is that in order to obtain a smooth and uniform film, it is crucial to start spinning immediately after applying the volatile solution to the substrate. Otherwise, the solution on the substrates will have higher viscosity and spread out less uniformly .
Although the surface roughness can be treated with annealing, nanopores observed inside propylamine-processed arsenic sulfide films are persistent to the high-temperature annealing [31, 44]. Such formation could degrade the quality of fabricated films by disrupting material homogeneity and introducing perturbation in refractive index, bandgap and roughness. In a recent pore study, an As and S vacancy coalescence mechanism was proposed to explain such process in the context of the dissolution chemistry, as discussed in 2.1 . Arsenic vacancies result from precipitation and sulfur vacancies arise from gas evolution above the threshold temperature around 120-130°C. One way to combat the pore problem while retaining wettability on common substrates is to add 10% EDA to the propylamine solution . EDA-dissolved solution/gel is molecular in nature and forms polymeric amorphous networks upon annealing [12, 23], without ammonium salt precipitation or gas evolution. Therefore, in solutions with 10% EDA, homogeneous films are obtained with no pores present (Fig. 3). As for practical applications, a more thorough removal of residual solvent is necessary to ensure high quality. This is true for almost all chalcogenide solution processes geared towards device fabrication.
4. Solution process in optical and photonic fabrication
4.1 Spin-coated waveguide over-cladding for low-temperature roughness reduction
Annealing is a common fabrication step to smooth out the surface roughness of chalcogenide structures, and it can be the key to reduce optical loss. However, some chalcogenides such as Ge-Sb-S are thermally less stable and prone to surface crystallization, which make them unsuitable for high-temperature annealing. To overcome this problem, the solution process can be used to deposit a arsenic sulfide over-cladding structure and achieve roughness reduction at sub-Tg annealing .
Carlie et al. demonstrated such surface roughness reduction in evaporated Ge-Sb-S structures while achieving optimization in refractive index, density and optical losses. Rib waveguides are first fabricated with Ge-Sb-S, onto which a 25mg/ml As-S solution is spin-coated and act as a waveguide over-cladding structure (Fig. 4). After heat treatment, the surface roughness is reduced from 50nm to ~5nm and the sidewall roughness is reduced from 19nm to 1.4nm. The optical loss measured at 1550nm also becomes smaller after spin-coating .
4.2 Thick film deposition and multilayer structures
Another advantage of the solution process over other conventional fabrication methods is the ability to deposit thick films. Such structures have technological importance in a number of applications, including data storage, plasmonics or high contrast photodarkening and photodoping in optoelectronics.
In a recent study, Zha et al. demonstrated 4µm single film deposition by spin-coating and over 10µm multilayer structures . Single layers are first deposited on salt substrates which can be entirely dissolved in water to give free-standing chalcogenide films. Multilayer structures are achieved by laminating these layers onto a spin-coated base film. The entire construct then undergoes an annealing step to remove liquid and any interfaces between the layers (Fig. 5). Homogeneous thick structures are fabricated using the same chalcogenide material, while heterogeneous multilayer structures can be created with layers of different solution-processed chalcogenides, or vacuum-coated metals. Photo-enhanced silver diffusion and refractive index increase over 0.2 are shown in the Ag-As2S3 multilayer structures .
4.3 Soft lithography and waveguide integration
Solution-based processing enables new techniques for waveguide integration that are unattainable by evaporation methods. Using the flow properties of liquids, micromolding in capillaries (MIMIC) [46, 47] and micro transfer molding (µTM) [47, 48] can directly integrate chalcogenide glass waveguides with existing optical devices without extra etching steps. The advantages of these techniques are multifold. The methods are simple and require no elaborate high-vacuum equipment. Moreover, complex geometries such as bend waveguides, y-splitters or interferometers can easily be realized using corresponding molds . The dimensions of the waveguides can be in the order of tens of microns in width and height, which makes them suitable for integration with mid-infrared devices such as quantum cascade lasers (QCLs). In comparison, these large dimensions are difficult to obtain by evaporation methods. Furthermore, the process requires only moderate temperatures, although the surface roughness and waveguide loss can be further reduced by a subsequent annealing step at higher temperatures such as 120°C for arsenic sulfide .
The MIMIC method uses a soft and flexible polydimethylsiloxane (PDMS) mold which has a reverse pattern from photolithography-made reusable master mold . After the PDMS mold is laid onto a substrate, a small amount of the chalcogenide solution is deposited at the channel entrance and fills up the channel by capillary action (Fig. 6). After baking, the PDMS mold can be easily removed. Thus, an arsenic sulfide waveguide is directly integrated with a QCL (Fig. 7) . For structures with a small cross section (< 5 µm in dimension), the MIMIC method is limited by the capillary forces to shorter filling lengths.
The µTM method uses similar PDMS molds and complements the MIMIC method, because it is not limited by capillary forces or the mold geometrics. In this method, the solution is spin-coated to form a thin film directly on the patterned PDMS mold. Immediately after spinning, a substrate is put on top of the film, and the whole assembly is transferred to a vacuum oven where a weight is added on top of the stack to increase the adhesion between film and substrate. The sample is baked at temperatures up to 100°C before the PDMS mold is removed, leaving behind a patterned glass film on a substrate, see Fig. 8 .
Additionally, there is another method involving solution processing for integration and is called capillary force lithography (CFL). In this case, a PDMS mold is pressed into a freshly spin-coated chalcogenide film which is then heated to temperatures as low as 150°C to effect viscous flow of the film. Grating patterns of 200µm line height and 2µm can be fabricated with a surface roughness of 0.9nm . Compared to conventionally prepared films, spin-coated films are moldable at lower temperatures.
4.4 Inkjet printing of microlenses
Another interesting application of solution-processing comes from the direct printing of microlenses. This technique enabled by inkjet-printing chalcogenide solutions can directly write patterns or precisely deposit individual microscopic infrared lenses. This method can find its application in places where the deposition surface is non-flat and becomes challenging for traditional methods. Inkjet printing overcomes the problem and allows deposition on any user-defined surfaces. Potentially, it also enables efficient lens fabrication in a large scale.
By controlling the solution concentration and pulse setting, spherical lenses of approximately 10-300µm in diameter are fabricated . Lens focal length ranges from 10 to 700µm. The nature of the deposition technique allows post-deposition modification of the lens. Additionally, thermal treatment can lead to an increase of the lens radius and refractive index, enabling precise adjustment of the F-number.
4.5 Chalcogenide layers in fibers
Depositing chalcogenide materials inside photonic crystal fibers is needed for supercontinuum generation and other non-linear applications. However, the hollow cores in such fibers are only a few microns in radius and would be difficult to achieve homogeneous deposition with the evaporation method. In such situation, solution process stands out for its deposition flexibility.
Marcos et al. chose butylamine and EDA solutions to pattern thin and thick As2S3 layers on the channel surfaces inside a photonic crystal fiber (Fig. 9). The solution fills the micron size channels under capillary force and the process may take hours depending on the solution and channel size. The optical transmission measurements reveals strong photonic bandgaps from visible to near-infrared wavelengths. From the transmittance spectra measured in the wavelength range 500–1750 nm, the lowest loss is found to be 3dB/cm .
4.6 Inverse opal photonic crystal
Lastly, inverse opal photonic crystal structures have been studied for their intriguing optical phenomena, as well as important applications in optical sensing, electronic paper and flexible laser devices. Given the 3d nature and complexity of this structure, it is desirable to choose the solution route over evaporation. Low-cost and low-temperature processing are the added advantages of this process.
Kohoutek et al. spin-coated arsenic sulfide solutions on highly-ordered silica colloidal crystal templates, filling in the interstitial spaces with solution. The template was subsequently rinsed in HF to remove the silica backbone and produce an inverse opal As-S photonic crystal (Fig. 10). Such structure demonstrated an enhanced reflectivity and wider photonic bandgap compared with the template film, opening the door to flexible colloidal crystal laser devices, photonic waveguide and novel chemical sensors .
We have shown a variety of applications in which the solution process is the vital step for chalcogenide deposition and structure fabrication. Its straight-forwardness and flexibility have proved ideal for fabricating convoluted structures or depositing materials on unconventional surfaces, where traditional methods become impractical. To date, the solution technique has been shown to effectively smooth out structure roughness, deposit thick layers, integrate with photonic components, print microlenses and fabricate complicated structures, while new applications continue to be developed. Where vacuum coating and other chalcogenide deposition methods have demonstrated useful applications in fibers, amplifiers and demultiplexers, solution based methods are just beginning to extend into these and other important areas, and show good potential for these applications . As the interests for chalcogenide device applications continue to expand in civil, medial and environmental sectors, the solution process will continue to provide unique opportunities.
This work is supported by NSF grant EEC-0540832 through the Mid-Infrared and Technologies for Health and Environment (MIRTHE) center.
References and links
2. C. Grillet, M. W. Lee, X. Gai, S. Tomljenovic-Hanic, C. Monat, E. Magi, D. J. Moss, B. J. Eggleton, S. Madden, D. Y. Choi, D. Bulla, and B. Luther-Davies, “Chalcogenide glass photonic crystals: progress and prospects,” Proc. SPIE 7609, 760913, 760913-9 (2010). [CrossRef]
3. B. J. Eggleton, B. Luther-Davies, and K. Richardson, “Chalcogenide photonics,” Nat. Photonics 5, 141–148 (2011).
4. A. Ganjoo, H. Jain, C. Yu, J. Irudayaraj, and C. G. Pantano, “Detection and fingerprinting of pathogens: Mid-IR biosensor using amorphous chalcogenide films,” J. Non-Cryst. Solids 354(19-25), 2757–2762 (2008). [CrossRef]
5. A. Kovalskiy, A. Vlcek, H. Jain, A. Fiserova, C. M. Waits, and M. Dubey, “Development of chalcogenide glass photoresists for gray scale lithography,” J. Non-Cryst. Solids 352(6-7), 589–594 (2006). [CrossRef]
6. J. J. Hu, V. Tarasov, A. Agarwal, L. Kimerling, N. Carlie, L. Petit, and K. Richardson, “Fabrication and testing of planar chalcogenide waveguide integrated microfluidic sensor,” Opt. Express 15(5), 2307–2314 (2007). [CrossRef] [PubMed]
7. V. Balan, C. Vigreux, and A. Pradel, “Chalcogenide thin films deposited by radio-frequency sputtering,” J. Optoelectron. Adv. Mater. 6, 875–882 (2004).
8. K. E. Youden, T. Grevatt, R. W. Eason, H. N. Rutt, R. S. Deol, and G. Wylangowski, “Pulsed-laser deposition of Ga-La-S chalcogenide glass thin-film optical wave-guides,” Appl. Phys. Lett. 63(12), 1601–1603 (1993). [CrossRef]
9. J. J. Hu, J. D. Musgraves, N. Carlie, B. Zdyrko, I. Luzinov, A. Agarwal, K. Richardson, and L. Kimerling, “Development of chipscale chalcogenide glass based infrared chemical sensors ,” Quantum Sensing and Nanophotonic Devices 7945,79452C (2011).
10. K. Richardson, L. Petit, N. Carlie, B. Zdyrko, I. Luzinov, J. Hu, A. Agarwal, L. Kimerling, T. Anderson, and M. Richardson, “Progress on the fabrication of on-chip, integrated chalcogenide glass (Chg)-based sensors,” J Nonlinear Opt Phys 19(01), 75–99 (2010). [CrossRef]
11. V. Takats, P. Nemec, A. C. Miller, H. Jain, and S. Kokenyesi, “Surface patterning on amorphous chalcogenide nanomultilayers,” Opt. Mater. 32(6), 677–679 (2010). [CrossRef]
12. V. Takáts, F. C. Miller, H. Jain, C. Cserhati, I. A. Szabo, D. Beke, and S. Kokenyesi, “Surface pattern recording in amorphous chalcogenide layers,” J. Non-Cryst. Solids 355(37-42), 1849–1852 (2009). [CrossRef]
13. S. Song, J. Dua, and C. B. Arnold, “Influence of annealing conditions on the optical and structural properties of spin-coated As2S3 chalcogenide glass thin films,” Opt. Express 18(6), 5472–5480 (2010). [CrossRef] [PubMed]
14. G. C. Chern and I. Lauks, “Spin-coated amorphous-chalcogenide films,” J. Appl. Phys. 53(10), 6979–6982 (1982). [CrossRef]
15. G. C. Chern, I. Lauks, and A. R. McGhie, “Spin coated amorphous chalcogenide films: Thermal properties,” J. Appl. Phys. 54(8), 4596 (1983). [CrossRef]
16. G. C. Chern and I. Lauks, “Spin coated amorphous chalcogenide films: Structural characterization,” J. Appl. Phys. 54(5), 2701 (1983). [CrossRef]
17. G. C. Chern, “Spin-coated amorphous chalcogenide films,” PhD thesis (University of Pennsylvania, 1984).
18. K. H. Norian, G. C. Chern, and I. Lauks, “Morphology and thermal properties of solvent-cast arsenic sulfide films,” J. Appl. Phys. 55(10), 3795–3798 (1984). [CrossRef]
19. G. C. Chern, I. Lauks, and K. H. Norian, “Spin-coated amorphous chalcogenide films: Photoinduced effects,” Thin Solid Films 123(4), 289–296 (1985). [CrossRef]
20. S. Mamedov, “On the macromolecular mechanism of dissolution of As2S3 films in organic solutions,” Thin Solid Films 226(2), 215–218 (1993). [CrossRef]
21. S. A. Zenkin, S. B. Mamedov, M. D. Mikhailov, E. Y. Turkina, and I. Y. Yusupov, “Mechanism for interaction of amine solutions with monolithic glasses and amorphous films in the As-S system,” Glass Phys. Chem. 23, 393–399 (1997).
22. J. Orava, T. Wagner, M. Krbal, T. Kohoutek, M. Vlcek, and M. Frumar, “Selective wet-etching of undoped and silver photodoped amorphous thin films of chalcogenide glasses in inorganic alkaline solutions,” J. Non-Cryst. Solids 352(9-20), 1637–1640 (2006). [CrossRef]
23. T. Kohoutek, T. Wagner, J. Orava, M. Krbal, A. Fejfar, T. Mates, S. O. Kasap, and M. Frumar, “Surface morphology of spin-coated As-S-Se chalcogenide thin films,” J. Non-Cryst. Solids 353(13-15), 1437–1440 (2007). [CrossRef]
24. T. A. Guiton and C. G. Pantano, “Sol-to-Gel and Gel-to-Glass Transitions in the As2S3-Amine system,” in Material Research Society Symposium Proceedings, (Materials Research Society, 1988), pp. 509–514. [CrossRef]
25. T. A. Guiton and C. G. Pantano, “Solution/gelation of arseic trisulfide in amine solvents,” Chem. Mater. 1(5), 558–563 (1989). [CrossRef]
26. T. Kohoutek, T. Wagner, M. Frumar, A. Chrissanthopoulos, O. Kostadinova, and S. N. Yannopoulos, “Effect of cluster size of chalcogenide glass nanocolloidal solutions on the surface morphology of spin-coated amorphous films,” J. Appl. Phys. 103(6), 063511 (2008). [CrossRef]
27. M. D. Michailov, S. B. Mamedov, and S. V. Tsventarnyi, “Dissolution Kinetics of Glassy Arsenic Sulfide in Alkali and Amine Solutions,” J. Non-Cryst. Solids 176(2-3), 258–262 (1994). [CrossRef]
28. M. D. Mikhailov and N. A. Anikina, “On the influence of stirring conditions on the dissolution rate of chalcogenide glasses and films,” Glass Phys. Chem. 37(1), 21–28 (2011). [CrossRef]
29. Y. Zou, H. Lin, O. Ogbuu, L. Li, S. Danto, S. Novak, J. Novak, J. D. Musgraves, K. Richardson, and J. Hu, “Effect of annealing conditions on the physio-chemical properties of spin-coated As2Se3 chalcogenide glass films,” Opt. Mater. Express 2(12), 1723–1732 (2012). [CrossRef]
30. T. Kohoutek, T. Wagner, M. Vlcek, M. Vlcek, and M. Frumar, “Spin-coated As33S67-xSex thin films: the effect of annealing on structure and optical properties,” J. Non-Cryst. Solids 352(9-20), 1563–1566 (2006). [CrossRef]
31. M. Waldmann, J. D. Musgraves, K. Richardson, and C. B. Arnold, “Structural properties of solution processed Ge23Sb7S70 glass materials,” J. Mater. Chem. 22(34), 17848 (2012). [CrossRef]
32. S. Song, “Solvent-casting of chalcogenide glasses and their applications in mid-infrared optics,” PhD thesis (Princeton University, 2011).
33. J. Orava, T. Wagner, A. Krbal, T. Kohoutek, M. Vlcek, and M. Frumar, “Selective wet-etching and characterization of chalcogenide thin films in inorganic alkaline solutions,” J. Non-Cryst. Solids 353(13-15), 1441–1445 (2007). [CrossRef]
34. J. Orava, T. Wagner, M. Krbal, T. Kohoutek, M. Vlcek, P. Klapetek, and M. Frumar, “Selective dissolution of Ag-x(As0.33S0.67-ySey)(100-x) chalcogenide thin films,” J. Non-Cryst. Solids 354(2-9), 533–539 (2008). [CrossRef]
35. T. Kohoutek, T. Wágner, M. Vlček, M. Vlček, and M. Frumar, “Physico-chemical properties of spin-coated Ag–As–Sb–S films,” J. Non-Cryst. Solids 351(27-29), 2205–2209 (2005). [CrossRef]
36. M. Krbal, T. Wagner, T. Kohoutek, P. Nemec, J. Orava, and M. Frumar, “The comparison of Ag-As33S67 films prepared by thermal evaporation (TE), spin-coating (SC) and a pulsed laser deposition (PLD),” J. Phys. Chem. Solids 68(5-6), 953–957 (2007). [CrossRef]
37. T. Kohoutek, T. Wagner, J. Orava, M. Frumar, V. Perina, A. Mackova, V. Hnatowitz, M. Vlcek, and S. Kasap, “Amorphous films of Ag-As-S system prepared by spin-coating technique, preparation techniques and films physico-chemical properties,” Vacuum 76(2-3), 191–194 (2004). [CrossRef]
38. J. J. Santiago, M. Sano, M. Hamman, and N. Chen, “Growth and optical characterization of spin-coated As2S3 multilayer thin-films,” Thin Solid Films 147(3), 275–284 (1987). [CrossRef]
39. S. Shtutina, M. Klebanov, V. Lyubin, S. Rosenwaks, and V. Volterra, “Photoinduced Phenomena in Spin-Coated Vitreous As2S3 and AsSe Films,” Thin Solid Films 261(1-2), 263–265 (1995). [CrossRef]
40. Y. Zha and C. B. Arnold, “Solution-processing of thick chalcogenide-chalcogenide and metal-chalcogenide structures by spin-coating and multilayer lamination,” Opt. Mater. Express 3(2), 309–317 (2013). [CrossRef]
41. J. Gutwirth, T. Wagner, T. Kohoutek, M. Vlcek, S. Schroeter, V. Kovanda, M. Vlcek, and M. Frumar, “Physical properties and structure of amorphous Ag-x(Sb0.33S0.67)(100-x) prepared by optically-induced diffusion and dissolution of silver into spin-coated amorphous Sb33S67 films and their application for optical recording,” J. Optoelectron. Adv. Mater. 5, 1139–1146 (2003).
42. T. Wágner, M. Krbal, T. Kohoutek, V. Peřina, M. Vlček, M. Vlček, and M. Frumar, “Kinetics of optically- and thermally-induced diffusion and dissolution of silver in spin-coated As33S67 amorphous films; their properties and structure,” J. Non-Cryst. Solids 326–327, 233–237 (2003). [CrossRef]
43. S. Song, N. Carlie, J. Boudies, L. Petit, K. Richardson, and C. B. Arnold, “Spin-coating of Ge23Sb7S70 chalcogenide glass thin films,” J. Non-Cryst. Solids 355(45-47), 2272–2278 (2009). [CrossRef]
44. N. Carlie, J. D. Musgraves, B. Zdyrko, I. Luzinov, J. J. Hu, V. Singh, A. Agarwal, L. C. Kimerling, A. Canciamilla, F. Morichetti, A. Melloni, and K. Richardson, “Integrated chalcogenide waveguide resonators for mid-IR sensing: leveraging material properties to meet fabrication challenges,” Opt. Express 18(25), 26728–26743 (2010). [CrossRef] [PubMed]
45. Y. Zha, S. Fingerman, S. J. Cantrell, and C. B. Arnold, “Pore formation and removal in solution-processed amorphous arsenic sulfide films,” J. Non-Cryst. Solids 369, 11–16 (2013). [CrossRef]
46. C. Tsay, E. Mujagić, C. K. Madsen, C. F. Gmachl, and C. B. Arnold, “Mid-infrared characterization of solution-processed As2S3 chalcogenide glass waveguides,” Opt. Express 18(15), 15523–15530 (2010). [CrossRef] [PubMed]
47. C. Tsay, F. Toor, C. F. Gmachl, and C. B. Arnold, “Chalcogenide glass waveguides integrated with quantum cascade lasers for on-chip mid-IR photonic circuits,” Opt. Lett. 35(20), 3324–3326 (2010). [CrossRef] [PubMed]
51. T. Kohoutek, J. Orava, T. Sawada, and H. Fudouzi, “Inverse opal photonic crystal of chalcogenide glass by solution processing,” J. Colloid Interface Sci. 353(2), 454–458 (2011). [CrossRef] [PubMed]